Max Robotics — Global Service Robot Directory + US Certification

About

David Hoeller is a robotics researcher whose work sits at the intersection of physics simulation, deep reinforcement learning, and legged robot locomotion. He has made foundational contributions to GPU-accelerated robot learning infrastructure, most notably through his involvement in **Isaac Gym** (322 citations) and the **Orbit framework** (226 citations), which have become widely adopted platforms enabling researchers worldwide to train robot policies at unprecedented speed and scale. His massively parallel deep reinforcement learning approach demonstrated that locomotion policies for real-world robots could be generated in minutes on a single GPU workstation, dramatically lowering the computational barrier for robotics research. Beyond simulation infrastructure, Hoeller has pushed the boundaries of what legged robots can physically achieve. His work on **ANYmal parkour** (216 citations) demonstrated fully learned agile navigation in complex environments, enabling quadrupedal robots to perform dynamic, contact-rich maneuvers previously considered beyond reinforcement learning's reach. He has further advanced end-to-end local navigation, traversability learning, and neural terrain reconstruction, creating a coherent research vision that connects simulation tooling to real-world deployment. Collectively, his papers have accumulated over 1,100 citations, reflecting significant influence on the field and establishing him as a leading voice in scalable, learning-based robotics.